Endoscopic administration of Coca-Cola for medical management of a wedged intestinal trichobezoar in a cat

Abstract

A 16-year-old, neutered male domestic longhair feline with gastroduodenal and jejunal trichobezoars was treated with administration of Coca-Cola through endoscopic injection catheter. Examination with a Karl Storz endoscope identified a trichobezoar causing suspected partial or early complete obstruction of the jejunum; however, the length of the scope was inadequate to retrieve the trichobezoar. Consequently, 55 mL of Coca-Cola and 5 mL of iohexol were instilled into an injection catheter to disrupt the trichobezoar. No peri- or postoperative complications were reported, and the cat recovered uneventfully.

Key clinical message:

This case report demonstrates a minimally invasive approach to treatment of a small intestinal trichobezoar in a cat with no intra- or postoperative complications. Coca-Cola infusion through an endoscopic injection catheter may be a viable treatment in cats when a surgical approach is not an option, although further cases are needed to determine whether these results can be generalized to the larger patient population.

Trichobezoars, or hair balls, are a common occurrence in domestic cats (1), particularly in long-haired cats living indoors (2). Overgrooming secondary to pruritic skin disease, flea infestation, or behavioral issues, and gastrointestinal motility disorders facilitate concentration of excessive hair within the gastrointestinal tract (1,3). The ingested hair is most often eliminated undigested in feces (1,2); however, intermittent vomiting is considered a physiologically normal response secondary to accumulation of hair in the stomach (3). Less commonly, hair balls become entrapped within the small intestine (3), resulting in partial or complete obstruction (1). They may also become lodged in the esophagus or nasopharynx after unsuccessful vomiting (4,5). Reported clinical signs associated with small intestinal trichobezoar obstruction include vomiting, inappetence, and abdominal discomfort (1,3), although few studies are available (1–3). Based on limited literature, exploratory laparotomy has been the primary treatment modality for small intestinal trichobezoar obstruction in cats (3).

In humans, Coca-Cola has been successfully used to dissolve gastric bezoars via nasogastric lavage and direct infusion into the bezoar (6,7). Most of these cases have involved phytobezoars, composed of nondigestible food materials (6–8). A single case of a gastric trichobezoar treated by Coca-Cola infusion and endoscopic fragmentation was reported (7), suggesting that Coca-Cola administration may be a safe, minimally invasive, and inexpensive method of treatment (7). However, other studies maintain that surgical removal is the most effective treatment modality despite the risk of postoperative complications, citing limited success with pharmacotherapy and endoscopic retrieval (9,10). To the authors’ knowledge, there are no published case reports of nonsurgically managed obstructive small intestinal trichobezoars in cats.

Case description

A 16-year-old, indoor-only, neutered male domestic longhair cat weighing 5 kg was presented to our emergency service clinic with a history of acute vomiting, inappetence, and lethargy. The owners reported that the vomiting began on the day before presentation, during which time the patient vomited approximately every 3 h. The patient’s last vomitus contained a small volume of liquid and a small trichobezoar. No defecation had been observed in the preceding 24 h. This senior cat was otherwise reported to be healthy, except for historical polyuria and polydipsia with some presumed age-related arthritis. Dietary indiscretion had also been reported in the past, with foreign material ingestion including paper and plastic. There was a suspicion that he had chewed a garbage bag 2 d before presentation, and there was concern for foreign material obstruction.

At presentation, the cat was fractious. Physical exam was limited but unremarkable, except for a body condition score of 2.5/5, temperature of 37.6°C, and ~7% dehydration. Abdominal palpation indicated moderate discomfort; however, palpation was limited due to patient temperament. Abdominal radiographs revealed a soft tissue material in the stomach as well as moderate gastric dilation. Small and misshapen kidneys were also noted, consistent with chronic renal disease and likely chronic renal infarcts. Abdominal Focused Assessment with Sonography for Trauma (AFAST) revealed moderate gastric dilation and hyperechoic, sharply shadowing material within the stomach. The pylorus and proximal duodenum were distended. In addition, AFAST identified bilaterally small and irregularly margined kidneys with reduced corticomedullary distinction. A complete blood (cell) count was unremarkable. Biochemistry revealed a moderate azotemia [BUN 23.0 mmol/L (reference range: 5.0 to 12.9 mmol/L), creatinine 219 μmol/L (reference range: 53 to 212 μmol/L)]; and urine specific gravity of 1.032 and proteinuria 1+ were noted. The owners elected to hospitalize with supportive care and evaluate progress as they had decided they would not take the cat to surgery. The cat was managed with intravenous fluids (Plasmalyte A; Baxter Corporation, Mississauga, Ontario) administered via constant rate infusion (CRI) at a rate of 25 mL/h; maropitant (1 mg/kg BW, IV, q24h); and pantoprazole (1 mg/kg BW, IV, q12h). Hydration improved, and repeat abdominal radiographs revealed a lobular soft tissue structure in the pyloric antrum with moderate gas dilation of the gastric fundus and cardia (Figure 1), concerning for a pyloric outflow obstruction. A single loop of small intestine in the central abdomen contained a mild amount of granular soft tissue material; the remaining small intestine was empty or contained a mild amount of fluid.

Figure 1.

Ventrodorsal and lateral radiographs of the abdomen of a 16-year-old neutered male cat. Lobular soft tissue material is lodged in the pylorus (arrow), with moderate gas dilation of the remainder of the fundus and cardia.

Surgical intervention was declined by the owners, but they elected to pursue endoscopic retrieval of the suspected foreign body under general anesthesia. The cat was sedated with butorphanol (0.2 mg/kg BW, IV) and midazolam (0.3 mg/kg BW, IV). Anesthesia was induced with alfaxalone (1 mg/kg BW, IV) and maintained with inhalant isoflurane.

Gastroduodenal endoscopy was carried out using a flexible GI scope (Karl Storz 60714NK; Karl Storz Endoscopy Canada, Mississauga, Ontario). A large trichobezoar fixed in the pyloric antrum (Figure 2 A) was successfully dislodged and retrieved endoscopically using rat tooth endoscopy forceps. Subsequent evaluation of the proximal duodenum was unremarkable, until the endoscope reached the distal duodenum/proximal jejunum. A very subtle mucosal lesion with some hair fibers was seen (Figure 2 B) and followed to an intestinal segment where a moderate amount of thick fluid (Figure 2 C) was present despite insufflation. The fluid was cleared with multiple sessions of irrigation and subsequent suction. In the lumen of the jejunum, another trichobezoar was appreciated (Figure 2 D). The endoscope was extended maximally to 140 cm; however, the trichobezoar was lodged beyond that point and retrieval attempts were unsuccessful, despite trying multiple snares, forceps, prongs, irrigation, and suction. It remained immobile despite external manipulation of the intestines. A weasel guide wire with over-the-wire balloon technique was not possible due to unavailability of fluoroscopic equipment.

Figure 2.

A — Trichobezoar lodged in the pyloric antrum of a 16-year-old cat, causing a gastric outflow obstruction seen on upper GI endoscopy. B — Subtle mucosal erosion with hair fibres seen in the duodenum. C — Duodenal segment with moderate amount of thick fluid present despite insufflation, suggesting possible obstruction. D — Trichobezoar lodged in intestinal segment that failed to move despite the use of endoscopic instruments.

As the owners were not willing to pursue surgery, a treatment used in humans with bezoars was recommended. The owners’ verbal consent was obtained before instilling Coca-Cola 55 mL + 5 mL nonionic, iodinated contrast (Omnipaque 350, iohexol injection USP, 76% w/v; GE Healthcare Canada, Mississauga, Ontario) through an endoscopic injection catheter to disrupt the trichobezoar. The endoscope was extended maximally to 140 cm before Coca-Cola infusion (Figure 3). Following Coca-Cola administration, on 2-view abdominal radiographs, there were gas bubbles in the stomach and small intestines (Figure 4). The iodinated contrast was added to identify any obstructive lesion within the gastrointestinal tract.

Figure 3.

Bubbles visible following endoscopic administration of Coca-Cola in a 16-year-old cat.

Figure 4.

Ventrodorsal and lateral radiographs of the abdomen of a 16-year-old cat after Coca-Cola administration. Radiopaque iodinated contrast material is present in transverse and proximal to mid-descending colon; there is granular soft tissue material mixed with the contrast material throughout the descending colon, which may represent fecal or foreign material.

The cat recovered without complications. The following day on physical exam, the abdomen was tense, but no pain response was elicited during palpation. Hydration was deemed adequate, and repeat abdominal radiographs indicated that the entire volume of iodinated contrast had reached the colon with some granular soft tissue material identified in the descending colon — perhaps either fecal or foreign material (Figure 4).

The stomach was empty, and the remaining small intestine contained a modest amount of both fluid and gas, with no evidence of mechanical obstruction. The cat was discharged the same day with mirtazapine (2 mg, PO, q24h, PRN); maropitant (16 mg at a dose of 1/4 tablet, PO, q24h for 2 d); and a hair ball remover containing white petrolatum, mineral oil, and soybean oil gel (Laxatone; Vétoquinol, Lavaltrie, Quebec), at a dose of 1/2 teaspoon daily for 3 d, as per manufacturers’ recommendations. A follow-up appointment was recommended but, since the cat was doing well, the owners elected not to follow up. A phone call 6 wk after the initial discharge confirmed that the cat had a large bowel movement 2 d after the initial discharge and the vomiting resolved post endoscopy. The owners had no concerns and had elected to follow up with the family veterinarian as needed. No follow-up records were available for review.

Discussion

Although published data are limited, trichobezoars are common in domestic cats (1). Trichobezoars occur when strands of hair that are swallowed evade gastric peristalsis and remain in gastric mucosal folds. Peristaltic contractions within the stomach drive formation of a hair ball as the volume of retained hair increases (11,12). Accumulation of hair may result in vomiting or, less frequently, obstruction of the small intestine, nasopharynx, or esophagus (3–5).

Bezoar composition, size, and location dictates the ideal treatment modality (10). In human medicine, endoscopic fragmentation and extraction have been used to successfully treat gastric bezoars, although surgical treatment via laparotomy or laparoscopy is often employed when more conservative methods fail (9,10,13). There are several case reports describing successful treatment of gastric bezoars in humans with nasogastric or endoscopic administration of Coca-Cola; however, these cases are usually restricted to phytobezoars (7,8). A 2013 review summarizing responses of phytobezoars to Coca-Cola administration reported that 91.3% of phytobezoar cases resolved with Coca-Cola instillation, either alone or in conjunction with endoscopy (14). Human literature indicates that the high density of large trichobezoars makes them resistant to endoscopic disruption; instead, they require surgical extraction (10,15). In a review of 40 reported cases of trichobezoars in humans, endoscopic retrieval of the trichobezoar was successful in only 2 cases, whereas the remaining cases were refractory to endoscopy and required laparoscopic surgery or laparotomy (16). A single study described trichobezoar treatment with Coca-Cola in humans, although the author noted that this method was not as successful for treating trichobezoars as for treating phytobezoars (7). No standard protocol for Coca-Cola administration exists (16), although success has been reported with peroral administration, direct intra-bezoar infusion, and nasogastric lavage (9,13,14). A recent report by Banse et al explored management of gastric and enteric persimmon phytobezoars in equids using oral or intragastric cola. Of equids treated with cola administration alone, 2 of 4 had resolution of the phytobezoars. The 2 that did respond received large volumes of cola (20 to 24 L/d), compared to < 4 L/d in those that did not respond. In all cases, cola was administered over several days, although treatment protocols varied (17).

The mechanism of action is unknown, but the ability of Coca-Cola to dissolve bezoars may be in part attributed to its mucolytic action imparted by bicarbonate, as well as the ability of carbon dioxide bubbles to digest fibers (13,14,18,19). Moreover, Coca-Cola is composed of phosphoric acid and carbonic acid and has a pH of 2.6, similar to gastric acid. The acidifying effects of Coca-Cola further digest the bezoar fibers (19). This mechanism is supported by evidence that low gastric acid promotes phytobezoar development, as do impaired peristalsis and poor pyloric function (20). Similar ingredients are also present in alternative Coca-Cola beverages, including Coca-Cola Light, Diet Coke, and Coca-Cola Zero Sugar. These Coca-Cola beverages have the same lytic action and ability to disrupt bezoars (16,21); however, it is unknown whether other carbonated beverages have the same potential (16).

To the authors’ knowledge, no published research has examined risks of Coca-Cola administration via endoscopy, although Coca-Cola is routinely administered in smaller volumes (< 5 mL) through an esophageal or a percutaneous endoscopic gastrostomy tube to break up food residues when the tube is clogged with food material. Coca-Cola is not approved for use in domestic species as a food or as a drug.

The pH of the healthy feline small intestine pre-feeding and post-feeding is 8.2 and 8.3, respectively, based on a continuous, noninvasive pH monitoring system (22). The decrease in pH caused by Coca-Cola administration may pose some risk to mucosal integrity (23), which is particularly relevant since the presence of a trichobezoar itself predisposes to mucosal injury and gastroduodenal ulceration (24). Coca-Cola consumption in humans has been associated with markedly decreased duodenal pH and decreased periods of intestinal alkalinization, increasing risk of duodenal ulceration (23). In rats, intraluminal perfusion of acetic acid (pH 2.4 to 2.6) induces diffuse intestinal mucosal damage marked by edema, neutrophil recruitment, and denudation of villi in a concentration-dependent manner (25). Similarly, oral administration of salicylic acid in rabbits induced morphological changes in the gastric and intestinal mucosa characterized by villus damage and altered mucous secretion (26).

In humans, carbonated drinks have been associated with various gastrointestinal disorders. Carbon dioxide is the ingredient most often linked with altered gastrointestinal physiology, through direct effects on mucosal integrity as well as the pressure exerted by gas (27). It is well-established that gas accumulation in the gastrointestinal tract causes abdominal distension and pain. In humans, luminal gas has been closely linked with anorexia, nausea, bowel dysfunction, eructation, and borborygmi, among other complaints (28). In dogs, aerophagia has been postulated as a risk factor for gastric dilatation-volvulus (29), further suggesting that accumulation of large volumes of gas in the small intestine is not a benign process. However, the degree of gas accumulation in veterinary species following acute intestinal administration is unknown.

In this case, Coca-Cola was administered directly into the small intestine, rather than into the stomach as described in the literature. Most xenobiotics are absorbed primarily in the small intestine due to its increased membrane permeability and surface area, although gastric motility can significantly affect the rate of drug absorption. The stomach’s effect on absorption is largely irrelevant for the purpose of this report, and the rate of Coca-Cola absorption can largely be controlled by rate of administration through the endoscope (30,31).

Although not explored in veterinary literature, contrast agents are frequently combined with carbonated drinks in human medicine without adverse effects. In fact, the US Food and Drug Administration data sheet for Omnipaque, a brand of iohexol injection, recommends mixing the contrast with carbonated beverages for oral administration in humans (32).

Repeat abdominal radiographs taken the day after the procedure had no evidence of excessive gas within the intestinal lumen, or of intestinal perforation, which would have been seen accompanied by leakage of contrast material into the abdominal cavity.

To the authors’ knowledge, all published cases of obstructive intestinal trichobezoar treatment in cats have involved surgical extraction via laparotomy or enterotomy (3). In many cases, however, endoscopy is a viable alternative to surgical retrieval of foreign bodies. This procedure is minimally invasive and provides the endoscopist an opportunity to assess mucosal damage, while avoiding postoperative pain and limiting recovery time. In a study exploring endoscopic retrieval of oesophageal and gastric foreign bodies in cats, 96.2% of cats survived endoscopic examination. Of cats that underwent endoscopic retrieval, 90% were hospitalized for ≤ 1 d, whereas only 6% were hospitalized for > 3 d (33). Major complications occurred in 15.4% of cats that underwent endoscopy, and included unsuccessful retrieval of foreign material, esophageal stricture formation, thoracic complications, and cardiopulmonary arrest (33). Although feline data are limited, the survival rate in cats undergoing a celiotomy after presenting with a gastrointestinal foreign body has been reported to be as low as 63% with linear foreign bodies (34). Although endoscopic retrieval of foreign bodies is not without risks, it avoids common surgical complications, including wound dehiscence or leakage of gastric contents causing peritonitis (33,35).

Limitations of this report include the small sample size (n = 1) and associated limited ability to generalize to the greater patient population. As well, since this case report presents information from uncontrolled observation, cause and effect cannot be established. Further, due to uncertainty relating to the degree of jejunal obstruction caused by the trichobezoar, it is difficult to evaluate the effects of treatment. We do believe that the trichobezoar was wedged, as most luminal material will move easily with irrigation, suction, or manipulation with endoscopic instruments during endoscopy, and this was not the case. Another limitation of this case report was the lack of follow-up data. Although abdominal radiographs were obtained following Coca-Cola administration, a repeat endoscopy was not done. In addition, no follow-up records were accessible. The limited information available makes it impossible to evaluate long-term complications of treatment and prevents us from drawing further conclusions about treatment efficacy in general.

This report presents a case of a small intestinal trichobezoar in an aged cat treated with direct endoscopic infusion of Coca-Cola. To the authors’ knowledge, there are no published case reports describing this treatment modality in cats. The peri-and postoperative risks associated with abdominal exploratory surgery and the lack of established research in this area warrants further exploration of alternative treatment modalities for small intestinal trichobezoars. CVJ